CN106092635A - A kind of mattress pressure comfort evaluation methodology - Google Patents

Abstract

The invention discloses a mattress pressure comfort evaluation method. The invention includes the following steps: (1) Installing a body pressure measuring instrument on the mattress to test the body pressure distribution of a person sleeping on the mattress to obtain body pressure distribution data ; (2) Mathematically process the body pressure distribution data to calculate the maximum pressure, average pressure, maximum pressure gradient, average pressure gradient, and contact area; (3) After the subjects had sleep experience on the mattress, they gave Evaluation data of mattress comfort; (4) Using GA‑Elman neural network to establish a coupling model between the subject’s body pressure distribution index data and the corresponding evaluation data, and output the evaluation results through the coupling model; GA‑Elman The neural network is to optimize the initial network weights of the Elman neural network structure by genetic algorithm. The present invention combines the body pressure distribution index and human body evaluation data to evaluate the comfort of the mattress, thereby helping people to choose a mattress with better comfort and ensuring people to enjoy comfortable sleep.

Description

A method for evaluating mattress pressure comfort

technical field

The invention relates to a method for evaluating the pressure comfort of a mattress.

Background technique

Mattress is the necessary furniture for the human body to sleep, and its comfort greatly affects the quality of people's rest. A comfortable mattress can relax the muscles of the waist, back and buttocks well. In 1986, a survey by Addison et al. showed that 7% of American adults stated that their sleep problems were related to the comfort of their own mattresses. Comfort is a very complex system, which is affected by human physiological and psychological factors, so it is difficult to study. Despite extensive research, controversy remains. This controversy is not only manifested in the definition of methods, but also in the evaluation of various evaluation methods. It can be said that the various evaluation methods of comfort have their own advantages and disadvantages, and there is no uniform standard. Therefore, it is of great significance to establish a scientific, comprehensive and objective evaluation system for the evaluation of sofa comfort.

In the prior art, most methods use body pressure distribution indexes to evaluate the comfort of mattresses, and the methods of calculating and selecting body pressure distribution indexes are also different. These methods are to let the subject lie on the mattress to be tested, use a body pressure tester to test the body pressure distribution, and perform mathematical processing on the obtained body pressure data to obtain the body pressure distribution index, which may be different, including the maximum Pressure, average pressure, maximum pressure gradient, average pressure gradient, etc. These indicators are used to evaluate comfort. This method can help people choose a mattress that is more suitable for them, improve sleep quality, and improve sleep caused by uncomfortable mattresses. Problem with poor quality. The existing method does not establish a quantitative model to evaluate the comfort of the mattress by using the body pressure distribution index, and the evaluation result is not very good; in addition, the existing method does not combine the subjective comfort evaluation data of the subjects , the comfort is evaluated only by the body pressure distribution index, and the reliability of the result is not high.

Contents of the invention

In order to improve the above problems, the present invention provides a method for evaluating the pressure comfort of a mattress.

In order to achieve the above object, the technical scheme adopted in the present invention is as follows:

A method for evaluating mattress pressure comfort, comprising the following steps:

(1) Install a body pressure measuring instrument on the mattress to test the body pressure distribution of a person sleeping on the mattress to obtain body pressure distribution data;

(2) Mathematically process the body pressure distribution data to calculate the maximum pressure, average pressure, maximum pressure gradient, average pressure gradient, and contact area;

(3) After the subjects had a sleeping experience on the mattress, they gave the evaluation data of the comfort of the mattress;

(4) The GA-Elman neural network is used to establish the coupling model between the subject's body pressure distribution index data and the corresponding evaluation data, and the evaluation results are output through the coupling model; the GA-Elman neural network is used to optimize the Elman Initial network weights for the neural network structure.

Further, in the step (3), the comfort evaluation data is represented by a number from 1 to 10, and the corresponding relationship is: 1 for comfort, 3 for slight discomfort, 5 for discomfort, 7 for extreme discomfort, and 9 for unbearable .

Further, in the step (4), the algorithm of the GA-Elman neural network is as follows:

a. Parameter setting, including individual coding string length L, population size M, termination genetic algebra G, crossover operator Pc, and mutation operator Pm;

b. Individual binary codes to generate initial populations;

c. Calculate the individual fitness value through the fitness function;

d. When the individual fitness value reaches the expected error, stop inheritance and go to step g, otherwise, go to step e;

e. Through genetic operations of selection, crossover and mutation, generate the next generation population;

f. The fitness value of the individual in the Gth generation reaches the expected error, and the algorithm terminates;

g. Individuals who reach the expected error are given the initial weight of the Elman of the present invention.

It is worth noting that the maximum pressure reflects the hardness of the mattress. When the type and structure of the mattress are determined, the maximum pressure of the harder mattress is higher, and the maximum pressure of the softer mattress is smaller and the maximum pressure is larger. At the same time, under the dual effects of large local pressure and time, the human body itself will also feel uncomfortable; the average pressure reflects the overall effect of the mattress on the human body to some extent. Among them, the smaller the value, the better the effect of the mattress on dispersing the weight of the human body, and the smaller the pressure stimulation of the mattress perceived by the human body; the mattress with higher stiffness also has a larger maximum pressure gradient, and the mattress with a more reasonable material distribution The gradient is small, and the human body feels more comfortable.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

The present invention combines the body pressure distribution index and human body evaluation data to evaluate the comfort of the mattress, thereby helping people to choose a mattress with better comfort and ensuring people to enjoy comfortable sleep.

Description of drawings

Fig. 1 is a flowchart of the present invention.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings and examples, and the embodiments of the present invention include but not limited to the following examples.

Example

A method for evaluating mattress pressure comfort, comprising the following steps:

(1) Use the body pressure distribution imaging system of XSENSOR company to collect the body pressure distribution data of the subjects. The XSENSOR pressure distribution test and analysis system can measure the pressure distribution of the human-bed contact surface, display the body pressure in real time through two-dimensional and three-dimensional color images, and can record and store the body pressure distribution data of the subjects. The steps of the body pressure distribution test experiment are as follows:

① Spread the body pressure distribution mat on the mattress, and the subjects lie flat on the mattress;

②Open the body pressure test software to collect data, observe the body pressure distribution map on the computer screen, and after the image is stable, close the body pressure test software and stop the collection;

③Record the body pressure distribution indicators calculated based on the collected data, including contact area, maximum pressure, average pressure, maximum pressure gradient, and average pressure gradient. Among them, the contact area is defined as the sum of the areas of all test points; the maximum pressure is the maximum value of all test points, which is: Pm=max (P1, P2, P3, ..., PN) N is the number of test points; the average pressure is all Arithmetic mean of the pressure at the pressure point (test point); the maximum pressure gradient is: Gm=max (gradG1, geadG2, ..., gradGN) N is the number of side points; the average pressure gradient is the arithmetic mean of the pressure gradient at each pressure point.

(2) According to the subjective comfort evaluation table, the subjects gave the subjective comfort evaluation data after the sleep experience on the mattress. The comfort evaluation table is represented by numbers from 1 to 10, and the corresponding relationship is as follows: Comfort 1 for mild discomfort, 3 for mild discomfort, 5 for severe discomfort, and 9 for unbearable discomfort.

(3) The CA-Elman neural network is used to establish the coupling model between the subject's body pressure distribution index data and the corresponding subjective comfort evaluation data. network weight. The steps of the algorithm are as follows:

a. Parameter setting, including individual coding string length L, population size M, termination genetic algebra G, crossover operator Pc, and mutation operator Pm;

b. Individual binary codes to generate initial populations;

c. Calculate the individual fitness value through the fitness function;

d. When the individual fitness value reaches the expected error, stop inheritance and go to step g, otherwise, go to step e;

e. Through genetic operations of selection, crossover and mutation, generate the next generation population;

f. The fitness value of the individual in the Gth generation reaches the expected error, and the algorithm terminates;

g. Individuals who reach the expected error are given the initial weight of the Elman of the present invention.

(4) The mean square error and correlation coefficient were used to evaluate the error of the mattress pressure comfort coupling model.

It is worth noting that, as shown in Figure 1, the present invention collects the body pressure distribution data by the body pressure distribution collection system, calculates the body pressure distribution index, and establishes the GA-Elman neural network model in combination with the subjective evaluation, and then the GA-Elman model Judgment is made, if the comprehensive data reaches the evaluation index, the pressure comfort evaluation result is output, otherwise, return to the beginning.

According to the above-mentioned embodiments, the present invention can be well realized.

Claims (3)

1. a mattress pressure comfort evaluation methodology, it is characterised in that comprise the following steps:

(1) arranging body pressure measuring instrument on mattress, test person is slept the pressure distribution on mattress, is obtained pressure distribution data；

(2) pressure distribution data are made Mathematical treatment, be calculated maximum pressure, mean force, maximum pressure gradient, mean pressure Power gradient, contact area；

(3), after subjects carried out sleep experience on mattress, the evaluating data of mattress comfortableness is given；

(4) use GA-Elman neutral net, set up between the pressure distribution achievement data of subjects and corresponding evaluating data Coupling model, and export evaluation result by coupling model；GA-Elman neutral net is for using genetic algorithm optimization Elman The initial network weights of neural network structure.

A kind of mattress pressure comfort evaluation methodology the most according to claim 1, it is characterised in that in described step (3), The evaluating data of comfortableness is represented by the number of 1～10, and corresponding relation is: comfortable is 1, and slight discomfort is 3, and discomfort is 5, the most not Fit is 7, it is difficult to standing is 9.

A kind of mattress pressure comfort evaluation methodology the most according to claim 2, it is characterised in that in described step (4), The algorithm of GA-Elman neutral net is as follows:

A, parameter set, and including individual UVR exposure string length L, group size M, terminate genetic algebra G, crossover operator Pc, variation calculation Sub-Pm；

B, individual binary coding, generate initial population；

C, by fitness function calculate ideal adaptation angle value；

D, ideal adaptation angle value reach anticipation error, stop heredity, go to step g, otherwise, go to step e；

E, by selecting, intersect and the genetic manipulation of variation, produce population of future generation；

F, G reach anticipation error for the fitness value of interior individuality, and algorithm terminates；

G, reach the individual initial weight giving Elman of the present invention of anticipation error.